目的　探讨肱骨头扭转角两种不同测量方法的准确性以及在肩关节置换中的应用。  方法　随机选取20例成人肱骨干标本（完整标本，无骨折及病理性损害），应用传统二维CT法及简易二维CT法测量肱骨头扭转角度，并与直接解剖测量所得数据进行统计学分析，验证两种测量方法的临床准确性，并指导其在肩关节置换中的应用。  结果　传统二维CT法测得数据为（32.1±13.9）°、简易二维CT法测得数据为（30.2±14.3）°、解剖测量法测得数据为（32.1±13.6）°。3组间行单因素方差分析，P>0.05，无显著统计学差异。行LSD-t检验分析，两组二维CT测量方法与直接解剖测量法比较P值分别为0.667、0.995，均大于0.05，无显著统计学差异。  结论　两种方法测得结果均能较准确反映实际扭转角度数，而简易二维CT法操作简便、直观、快捷，实用性更强。肱骨头扭转角的个体差异较大，术前准确评估其数值，对患者的个性化肩关节置换手术有重要意义。

Objective　To verify the accuracy of humeral head retroversion angle with  two different methods commonly used in clinical practice and assess the value in the shoulder arthroplasty.　Methods　20 dry humerus specimens (bones with closed growth plates and no signs of degenerative or post-traumatic alterations. No information about sex or age) were scanned with 64 slice spiral CT, using two methods-the traditional two-dimensional CT measurement and a simple two-dimensional CT measurement, to measure the humeral head retroversion angle; then use the anatomical measurement to obtain the standard data; At last, the three groups data on humeral head retroversion angle were statistically analyzed.　Results　The angle of retroversion was （32.1±13.9）° using  traditional two-dimensional CT method, (30.2±14.3)° using the simple two-dimensional CT method, and (32.1±13.6)° using anatomical dissection. There was no statistical difference between groups by statistics (P>0.05).　Conclusion　Both methods can assess the humeral head retroversion angle very accurately, but the new two-dimensional CT method is relatively simpler, faster, more intuitive and practical. The Humeral head retroversion angle has great individual differences, and it is of great significance to accurately assess its value before the individually surgery.